Discrepancies between population-based data and adverse reaction reports in assessing drugs as causes of acute pancreatitis

Background: Many drugs are believed, clinically, to cause acute pancreatitis. We used information held on the UK General Practitioner Research Database to compare risks for drugs for which reports of pancreatitis were common or uncommon. Methods: Drug prescriptions were examined in 3673 patients with acute pancreatitis and in matched controls. Odds ratios were calculated for recent (1-90 days before the episode), past (91-360 days before the episode) or continuing (prescription in both periods) use. Results: Odds ratios were markedly increased for recent antisecretory use in non-ulcer patients only [all H-2 -antagonists, 12.4 (9.5-16.4); all proton pump antagonists, 9.3 (6.6-13.0)], with smaller increases for past [3.1 (2.5-3.7) and 3.5 (2.6-4.6), respectively] and continuing [2.6 (2.2-3.1) and 3.7 (2.9-4.7), respectively] use in patients without ulcer. Recent users of mesalazine showed a markedly increased risk [9.0 (1.8-44.6)], with smaller increases in past and continuing users [4.5 (1.3-16.0) and 2.5 (1.2-5.0), respectively]. Odds ratios for other drugs, suspect or not, were modestly increased, irrespective of whether the use was recent, past or continuing. The presence of gall-stones was not associated with a modified risk. Conclusions: Mesalazine, azathioprine and antisecretory drugs in non-ulcer subjects may increase the risk of pancreatitis, but warnings of drug-induced pancreatitis are generally not accompanied by increased population risks.